Primary sourcelmarena.ai
View sourceEvent ·Multi-source
Gemini 3.8 Flash (High) 重塑 Agent Arena 前沿
GoogleDeepMind 发布的 Gemini 3.8 Flash (High) 在 Agent Arena 中排名第 14 位,净提升 +5.94%,任务成本为 0.22 美元。该模型在 Text Arena 中排名第 7 位,得分 1494 分,超越 Claude Opus 5 (High)。相比 Gemini 3.7 Flash (High),在多个类别中排名显著提升,如写作类别从第 7 名升至第 3 名。
Current read
GoogleDeepMind 新发布的 Gemini 3.8 Flash (High) 在多个竞技场表现优异,成本比同类模型低 44-50%。
11 sources78° AI heatLast updated 2026/9/2 22:16:14
Evidence
11 sourcesRelated source 1shao__meng
View sourceRelated source 2IT之家
View sourceRelated source 3elvis
View sourceRelated source 4Paul Couvert
View sourceRelated source 5Augment Code
View sourceRelated source 6AI Will
View sourceRelated source 7向阳乔木
View sourceRelated source 8Google DeepMind
View sourceRelated source 9Logan Kilpatrick
View sourceRelated source 10Patrick Loeber
View sourceConflict check
Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.